c Copernicus Gesellschaft 2001 Proceedings of ICRC 2001: 944
ICRC 2001
The charge ratio of atmospheric muons below 1.0 GeV/c: Status and Perspective I. M. Brancus1 , J. Wentz2 , B. Vulpescu1,3 , H. Rebel2 , A. F. Badea1 , A. Bercuci1,2 , H. Bozdog1 , M. Duma1 , H. J. Mathes2 , M. Petcu1 , C. Aiftimiei1 , and B. Mitrica4 1
IFIN-HH Bucharest, Romania Forschungszentrum Karlsruhe, Institut f¨ur Kernphysik, Germany 3 Universit¨at Heidelberg, Physikalisches Institut, Germany 4 Faculty of Physics, Bucharest University, Romania
Abstract. The compact WILLI device, built in IFIN-HH Bucharest (44◦ 26’N latitude, 26◦ 04’E longitude and 85 m a.s.l.) at a vertical cut-off rigidity of 5.6 GV, has been used for measurements of the charge ratio in the vertical atmospheric flux at momenta below 1 GeV/c. In this low energy range the studies of muon charge ratio provide information useful for the discussion of the so-called atmospheric neutrino anomaly and for studies of atmospheric neutrino and antineutrino fluxes. The experimental method is based on the observation of the reduced effective lifetime of negative muons, stopped in matter, as compared to the lifetime of positive muons. Avoiding the difficulties and the systematic errors of magnetic spectrometers, results with high accuracy are obtained, indicating a decrease of the muon charge ratio from 1.30 (at 0.87 GeV/c) to 1.15 (at 0.24 GeV/c). The detector WILLI has been recently transformed in a rotatable device for measuring the muon charge ratio for different angles of incidence in azimuthal and zenithal range. With such perspective a systematic “muon charge ratio spectroscopy” may provide interesting geophysical observations.
1
2.2
0.4 < Rc < 3.0 GV
CAPRICE 94 Rc = 0.5 GV Lynn Lake
3.0 < Rc < 5.8 GV
CAPRICE 97 Rc = 4.2 GV Fort Sumner
2
6.4 < Rc < 15.6 GV
1.8
1.6
1.4
1.2
1
0.8 1
10
pµ [GeV/c]
Fig. 1. Compilation of measurements on the charge ratio of vertical muons at low muon momenta. The different symbols stand for different intervals of the geomagnetic cut-off Rc at the location of the experiment (Vulpescu et al., 1998; Kremer et al., 1999).
Introduction
At low energies the actual interest in studies of the muon charge ratio arises from the aspects related to the so-called atmospheric neutrino anomaly and from examining the effects of the Earth’s magnetic field on the propagation of secondary cosmic rays (Ryazhskaya, 1996). When the primary cosmic radiation interacts with the Earth’s atmosphere, baryons and mesons are produced. Atmospheric muons originate mainly from the decay of charged pions and kaons in muons and the subsequent decay of the muons: π ± → µ± + νµ (νµ ) K ± → µ± + νµ (νµ )
R(µ+/µ-)
2
99.99 % 63.51 %
τ = 26 ns τ = 12 ns
Correspondence to: I. M. Brancus (
[email protected])
Muons have a relative large lifetime τ0 = 2.2 µs, thus only part of them decay: µ+ → e+ + νe + νµ µ− → e− + νe + νµ . Considering the decay chains, it is obvious that the ratio of positive to negative atmospheric muons, called the muon charge ratio: Rν =
µ+ µ−
maps the neutrino production and carries information on the hadronic interactions, used in the calculations of atmospheric
945
a)
WILLI-97 100 cm
100 c
WILLI-98/99
b)
m Trigger modules
87 cm
101 cm
Anti-counters
78 cm
20 x ( 1 cm Pb + 3 cm plastic scintillator + 1 cm Al )
200 cm
10 cm Pb e/γ shielding
Active modules
Fig. 2. a) The original sampling calorimeter configuration of the WILLI detector used in WILLI-97 runs. b) The second configuration, improved for the measurements on the muon charge ratio used in WILLI-98 and WILLI-99 runs.
neutrino fluxes. Super-Kamiokande (Fukuda et al., 1998) and other experiments find that the ratio of muonic to electronic neutrinos is much smaller than the theoretical predictions. In addition, the effect is depending on the angle of incidence of the neutrinos. A possibility to verify the invoked models, which are used for the neutrino flux calculations, is to compare the theoretical results for the muon flux, with the experimental data (Wentz et al., 2001). The compilation of muon charge ratio values measured in the last decades shows that at higher energies all experiments + are consistent with a value µµ− ≈ 1.3 (Fig. 1). At energies lower than 1.0 GeV, the experimental uncertainties appear to be larger and the data disagree with each other. This finding led us to perform measurements in this energy range.
2 The method and the apparatus An experiment have been developed for studying the charge ratio of atmospheric muons with an electromagnetic calorimeter WILLI (Weakly Ionizing Lepton Lead Interactions) set-up in IFIN-HH Bucharest, related to studies of the KASCADE experiment. Differently from other experiments, using magnetic spectrographs, in which the charged particle trajectories are measured before and after traversing a magnetic field, our method (Vulpescu et al., 1998) to determine the muon charge ratio is based on the different behavior of positive and negative muons stopped in matter. While positive muons decay with their natural life time, the negative muons are captured in atomic orbits and form muonic atoms, leading to a shorter lifetime of negative muons in matter. The WILLI device measures the effective life-time of the stopped
muons for both charge states by observing the appearance of decay electrons and positrons after the muon stopped. The initial configuration, WILLI-97 (Fig. 2a), consisted of 20 modules, each made of a lead plate of 1 m2 area, 1 cm thick and a plastic scintillator of 90 x 90 x 3 cm3 , contained in an aluminum box (1 cm thick lateral walls and bottom and 1 mm cover) (Vulpescu et al., 1998). In a further step (Fig. 2b), the detector has been optimized for the measurement of the muon charge ratio by removing the lead plates and improving the geometry in order to have a background rejection using four scintillator modules in vertical position as anti-counters to cover the sides of the detector (Vulpescu, 1999). With this configuration two sets of measurements have been performed with the detector placed at the basement of the Nuclear Physics Department under 60 cm reinforced concrete (WILLI-98) and in another building with less shielding material above, where data have been taken for muons at lower energy, WILLI-99 (Vulpescu et al., 2001). The detector is equipped with 40 photomultipliers providing energy deposit (amplitude, by 40 anodes and 40 dynodes) and time information (by 40 dynodes, from which 32 are analyzed, 28 signals of 14 layers and 4 signals of anti-counters). The trigger is made out of four separate signals corresponding to two of the detector layers which define the accepted solid angle of the device. The light of the scintillator is collected by two photomultipliers with wavelength shifters. The energy signals taken from the anode and the third dynode are conducted to an ADC. The timing signals, taken from the dynodes are analyzed by a Multiple Time Digital Converter. The signature of a stopped and decaying muon is a particle triggering the telescope, but not penetrating till the bottom of
1
10 3
2
10 4
Events
Events
946
WILLI-97 simulation 1 - µ+2 - µ- in scintillator 3 - µ- in aluminum 4 - µ in lead
10 3
1
WILLI-98 simulation 1 - µ+2 - µ in scintillator 3 - µ- in aluminum
2 10 2
3
3
4 2
3
4
5
0
t [µs]
2
4
6
Events
1
Events
0
WILLI-97 DATA line - free decay of µ+
8
10
t [µs]
WILLI-98 DATA line - free decay of µ
+
10 4 10 3
10 4
0
1
2
3
4
5
0
t [µs]
2
4
6
8
10
t [µs]
Fig. 3. Upper figures: Monte Carlo simulation results on the contribution of different absorber materials to the total decay curve, assuming N (µ+ )/N (µ− ) = 1. Lower figures: The experimental decay curves compared with the free decay of positive muons.
3
Monte Carlo simulations and results
The total decay curve of all muons measured in the detector is a superposition of several decay laws : dN N0 1 −t 1 −t = [Rc0 exp( ) + (Σcj ρj exp( ))] dt (R + 1) τ0 τ0 τj τj + R( µµ− )
N+ N−
positive muons. One notices that aluminum provides the best discriminating effect, reducing significantly the effective mean life time of negative muons as compared to that of positive muons, and the decay probability for Al is 39.05 % while for Pb only 2.75 %. It is seen also the improvement by removing the lead absorbers in the new configuration, resulting in an increase of the number of the detected muons which are stopped in aluminum. R(µ+/µ-)
the scintillator stack, together with the appearance of a delayed particle in the surrounding of the stopping locus. From the time difference between the incoming muon and the decay electron, the spectrum of the decay times is registered.
CAPRICE 94 Rc = 0.5 GV Lynn Lake
1.6
WILLI Rc = 5.6 GV Bucharest
CAPRICE 97 Rc = 4.2 GV
Fort Sumner 1.5
+
= represents the muon charge ratio, N , where N − being the number of positive and negative muons, respectively, N0 = N + + N − , ρj gives the decay probability, and τj indicating the mean lifetime of µ− with index j for different absorber materials, index 0 standing for positive muons. The expression contains four detector dependent constants cj , for WILLI-97 and three constants for WILLI-98/99, accounting for the stopping power in the materials and the detection efficiencies, given by the detector geometry, laboratory walls, thresholds, and angular acceptance, which have been determined by extensive detector simulations using the code GEANT (CERN, 1993). The charge ratio of the muons is estimated by fitting the decay law above to the experimental data. Fig. 3 shows the results of the simulations indicating the decay exponentials for different materials and the comparison of the experimental decay curve with the free decay of
1.4
1.3
1.2
1.1
1
0.9 -1 10
1
10
10
2
pµ [GeV/c]
Fig. 4. The WILLI results for the muon charge ratio at sea-level together with recent results from CAPRICE experiment for two locations with different geomagnetic cut-off rigidities (Kremer et al., 1999).
947 4
Fig. 5. A photograph of the modified WILLI detector for measuring muons with different angles-of-incidence.
The Tab. 1 presents the results of all three experiments with different detector configurations. In the WILLI-99 experiment the upper and the bottom sections of the detector have been analyzed separately. < pµ > 0.87 GeV/c 0.52 GeV/c 0.28 GeV/c 0.24 GeV/c 0.20 GeV/c
< θµ > 26◦ 19◦ 16◦ 17◦ 19◦
+
R( µµ− ) 1.30 1.27 1.21 1.18 1.15
∆R 0.05 0.01 0.02 0.02 0.02
WILLI 1997 1998 1999-lower half 1999-full 1999-upper part
Table 1. The results of the muon charge ratio measurements WILLI-97, WILLI-98 and WILLI-99 at sea level, pµ and θµ stand for the momentum and zenith angle of the muons.
Concluding remarks
The measurements of the muon charge ratio by means of magnetic spectrometers are affected by systematic effects at low muon energies due to problems in the particle and trajectory identification, leading to quite different results. Our method overcomes such difficulties by measuring the lifetime of muons stopped in matter. The measurements indicate a muon charge ratio value of 1.3 at a momentum of 0.8 GeV/c, decreasing to 1.15 at 0.2 GeV/c, what can be attributed to the geomagnetic cut-off. Our results are different from the CAPRICE results for Fort Summer in New Mexico where the geomagnetic cut-off is similar to that in Bucharest. The WILLI detector is a suitable instrument for the further investigation of the modulation of the muon charge ratio and therefore of the neutrino fluxes by the geomagnetic cutoff. The details about the geomagnetic influence have not been systematically explored using the observation of muons with different arrival directions. The WILLI detector is now modified in a rotatable set-up (Fig. 5), which will allow precise measurements of the East-West effect, caused by the anisotropy of the primary proton flux and the local magnetic field, bending charged particles on their way through the atmosphere. Acknowledgements. The authors affiliated to IFIN-HH Bucharest are grateful for the support received from the International Bureau Bonn, by WTZ-Project RUM-14-97 with the Romanian Ministry for Education and Research and from the Institut f¨ur Kernphysik, Forschungszentrum Karlsruhe. Parts of the work have been supported by the grant FZK I/72-709 of Volkswagen-Stiftung and by the grant WE 2426/1-1 by the Deutsche Forschungsgemeinschaft.
References Fig. 4 shows the results of WILLI compared to other recent experiments (Kremer et al., 1999). The WILLI results indicate a smooth decrease of the charge ratio towards lower energy, which could be expected due to geomagnetic cutoff. The CAPRICE experiment reported a similar decrease, but with lower values, for New Mexico, at nearly the same geomagnetic cut-off as in Bucharest. The CAPRICE results for Lynn Lake (where actually no cut-off effect is expected) show higher values for lower energy, but with a strange modulation around pµ = 0.8 GeV/c.
CERN Program Library Long Writeups W5013, 1993. Y. Fukuda et al., Phys. Rev. Lett. 81, 1562–1567, 1998; Phys. Lett. B 433, 9–18, 1998; Phys. Lett. B 436, 33–41, 1998. J. Kremer et al., Phys. Rev. Lett. 83, 4241–4244, 1999. O.G. Ryazhskaya, Il Nuovo Cim. 19C, 655, 1996; B. Vulpescu et al., Nucl. Instr. and Meth. A414, 205–217, 1998. B.Vulpescu, Report FZKA 6368, Forschungszentrum Karlsruhe, 1999. B. Vulpescu et al., J. Phys. G.: Nucl. Part. Phys. 27, 977–991, 2001. J. Wentz et al., J. Phys. G. Nucl. Part. Phys, in press; these proceedings
